Organometallics 2006, 25, 3301-3304
3301
A New Twist on Pincer Ligands and Complexes Liqing Ma,† Robert A. Woloszynek,† Weizhong Chen,‡ Tong Ren,‡ and John D. Protasiewicz*,† Departments of Chemistry, Case Western ReserVe UniVersity, CleVeland, Ohio 44106, and UniVersity of Miami, Coral Gables, Florida 33146 ReceiVed March 21, 2006 Summary: A new class of palladium pincer complexes bearing m-terphenyl scaffolds haVe been synthesized and structurally characterized. As a result of integrating phenyl rings into the chelating arms of m-xylyl type pincer ligands, highly twisted and rigid structures haVe been achieVed. Pincer ligands1,2 are an exciting class of ligands that are receiving increasing attention for applications ranging from catalysis and sensors to materials science.3 The name bestowed upon these ligands3f reflects their tenacious and terdentate binding nature. If one considers the most well-known pincer ligand platform, the m-xylyl framework [2,6-(ECH2)2C6H3]- (an ECE pincer, E ) donor atoms or groups such as NR2, PR2, OR, etc.; Scheme 1, top), one can envision how the presence of a formal negative charge and two exemplary five-membered chelate rings can impart great stability to such complexes. More recently, related pincer ligands have been constructed featuring N-heterocyclic carbenes as C-donors (ECE and CEC types) that vastly increase the diversity of this ligand class. Many of these systems are also designed such that the central, “anchoring” donor atom is often part of a planar group (such as a phenyl, pyridine, or N-heterocyclic carbene group). It has been recognized that if this ring is not in the plane containing the metal and the two outer donor atoms, then a twisted conformation is realized. The C2-symmetric and, hence, chiral nature of these complexes offers the potential of resolution and use in performing catalytic enantioselective transformations. Interconversion between these two atropisomers (Scheme 1, a and a′), however, prevents isolation of individual enantiomers, resulting in an “averaged” planar structure (Scheme 1, a′′). Highly twisted structures have been produced,4 and configurational stability for a twisted 2,6-lutidinyl-bis(carbene) complex was maintained up to 80 °C.4f It should be mentioned that chiral pincer ligands have been reported that include modifications to the methylene carbons or the use of stereogenic centers at the donor atoms of this versatile motif.5 In this report we demonstrate not only the * To whom correspondence should be addressed. E-mail: protasiewicz@ case.edu. Fax: (+1) 216-368-3006. † Case Western Reserve University. ‡ University of Miami. (1) Moulton, C. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1976, 1020. (2) (a) Rietveld, M. H. P.; Grove, D. M.; van Koten, G. New J. Chem. 1997, 21, 751. (b) Albrecht, M.; van Koten, G. Angew. Chem., Int. Ed. 2001, 40, 3750. (c) van der Boom, M. E.; Milstein, D. Chem. ReV. 2003, 103, 1759. (d) Crabtree, R. H. Pure Appl. Chem. 2003, 75, 435. (e) Slagt, M. Q.; van Zwieten, D. A. P.; Moerkerk, A.; Gebbink, R.; van Koten, G. Coord. Chem. ReV. 2004, 248, 2275. (3) (a) Ohff, M.; Ohff, A.; van der Boom, M. E.; Milstein, D. J. Am. Chem. Soc. 1997, 119, 11687. (b) van der Boom, M. E.; Liou, S. Y.; BenDavid, Y.; Shimon, L. J. W.; Milstein, D. J. Am. Chem. Soc. 1998, 120, 6531. (c) Jensen, C. M. Chem. Commun. 1999, 2443. (d) Sundermann, A.; Uzan, O.; Milstein, D.; Martin, J. M. L. J. Am. Chem. Soc. 2000, 122, 7095. (e) Morales-Morales, D.; Redon, R.; Yung, C.; Jensen, C. M. Chem. Commun. 2000, 1619. (f) Zhu, K. M.; Achord, P. D.; Zhang, X. W.; KroghJespersen, K.; Goldman, A. S. J. Am. Chem. Soc. 2004, 126, 13044. (g) van Koten, G. Pure Appl. Chem. 1989, 61, 1681.
Scheme 1. Typical “Pincer” Binding and Atropisomers
attainment of the highest twist angles to date but also systems having a very high degree of nonfluxionality and versatility. Our strategy was inspired by our past success in employing m-terphenyls to stabilize various materials having low-coordinate phosphorus atoms.6 Specifically, we sought to reposition phosphorus atoms on a terphenyl framework, leading to a new class of pincer ligands (Scheme 2). Our initial efforts, however, were somewhat disappointing in that ligands of the form [2,6(2-R2PCH2C6H4)2C6H4] did not undergo cyclometalation and pincer complex formation upon reaction with Pd(II) salts (unlike the case for [(2,6-(R2PCH2)2C6H4] ligands) but, instead, yielded complexes bearing trans-spanning diphosphines (Scheme 2, left).7 Using a common workaround, installation of a more reactive halogen atom in place of a hydrogen atom at the critical (4) (a) Grundemann, S.; Albrecht, M.; Loch, J. A.; Faller, J. W.; Crabtree, R. H. Organometallics 2001, 20, 5485. (b) Diez-Barra, E.; Guerra, J.; LopezSolera, I.; Merino, S.; Rodriguez-Lopez, J.; Sanehez-Verdu, P.; Tejeda, J. Organometallics 2003, 22, 541. (c) Lee, H. M.; Zeng, J. Y.; Hu, C. H.; Lee, M. T. Inorg. Chem. 2004, 43, 6822. (d) Tulloch, A. A. D.; Danopoulos, A. A.; Tizzard, G. J.; Coles, S. J.; Hursthouse, M. B.; Hay-Motherwell, R. S.; Motherwell, W. B. Chem. Commun. 2001, 1270. (e) Nielsen, D. J.; Cavell, K. J.; Skelton, B. W.; White, A. H. Inorg. Chim. Acta 2002, 327, 116. (f) Danopoulos, A. A.; Tulloch, A. A. D.; Winston, S.; Eastham, G.; Hursthouse, M. B. Dalton Trans. 2003, 1009. (5) (a) Xia, Y.-Q.; Tang, Y.-Y.; Liang, Z.-M.; Yu, C.-B.; Zhou, X.-G.; Li, R.-X.; Li, X.-J. J. Mol. Catal. A: Chem. 2005, 240, 132. (b) Motoyama, Y.; Nishiyama, H. Synlett 2003, 1883. (c) Medici, S.; Gagliardo, M.; Williams, S. B.; Chase, P. A.; Gladiali, S.; Lutz, M.; Spek, A. L.; van Klink, G. P. M.; van Koten, G. HelV. Chim. Acta 2005, 88, 694. (d) Williams, B. S.; Dani, P.; Lutz, M.; Spek, A. L.; van Koten, G. HelV. Chim. Acta 2001, 84, 3519. (e) Takenaka, K.; Minakawa, M.; Uozumi, Y. J. Am. Chem. Soc. 2005, 127, 12273. (f) Steurer, M.; Tiedl, K.; Wang, Y. P.; Weissensteiner, W. Chem. Commun. 2005, 4929. (g) Gerisch, M.; Krumper, J. R.; Bergman, R. G.; Tilley, T. D. Organometallics 2003, 22, 47. (6) (a) Shah, S.; Protasiewicz, J. D. Coord. Chem. ReV. 2000, 210, 181. (b) Smith, R. C.; Protasiewicz, J. D. Eur. J. Inorg. Chem. 2004, 998. (c) Smith, R. C.; Protasiewicz, J. D. J. Am. Chem. Soc. 2004, 126, 2268.
10.1021/om060256z CCC: $33.50 © 2006 American Chemical Society Publication on Web 06/01/2006
3302 Organometallics, Vol. 25, No. 14, 2006
Communications
Scheme 2. Relationship between Previous Trans-Spanning (Left) and New Pincer Complexes (Right)
Scheme 3. Pincer Ligand Synthesis Figure 1. ORTEP drawing (30% probability ellipsoids) of the molecular structure of 3a. Hydrogen atoms are omitted for clarity. Select bond lengths (Å) and angles (deg): Pd-P(1), 2.3071(7); Pd-P(1A), 2.3071(7); Pd-C(1), 2.067(4); Pd-Br, 2.5024(5); P(1)-Pd-P(1A), 169.48(4); C(1)-Pd-Br, 180.0.
C-1 position of the central ring afforded ligand precursors that readily reacted with low-valent complexes to provide the new C2-symmetric pincer complexes 3 and 4 (Scheme 2, right). The synthesis of the ligand precursors [2,6-(2-R2PCH2C6H4)2C6H3Br] (1a-c) is outlined in Scheme 3. It was discovered that the shorter synthesis used to prepare the nonhalogenated
analogues [2,6-(2-R2PCH2C6H4)2C6H4] had to be modified. Specifically, the key intermediate [2,6-(2-BrCH2C6H4)2C6H3Br] (v) could not be cleanly prepared by direct bis-monobromination of the tolyl groups of i, due to the formation of mixtures of mono-, bis-, tri-, and tetrahalogenated products that were difficult to separate.8 Instead, synthesis of the readily purified ii allowed facile synthesis of iii and iv, which could then be used to access v from i in 56% overall yield. From precursor v the three new diphosphines 1a-c were isolated in good yields. Each material is readily characterized by standard 1H and 31P NMR spectroscopy.9 Notably, the diphosphine 1a shows two 31P NMR signals in its spectrum, consistent with the presence of two isomers (syn and anti) in about a 50:50 ratio. 31P NMR spectra of diphosphines 1b,c indicate a much smaller amount (ca. 10% or less) of a second isomer. The increased steric bulk of the benzyl groups in 1b,c presumably disfavors the syn isomers, as the precursors i-v all show syn and anti isomers in about a 50:50 ratio (by 1H NMR spectroscopy). Nevertheless, diphosphines 1a-c react with Pd2(dba)3 to quantitatively yield pincer complexes 3a-c (Scheme 2, right), as indicated by new signals ca. 25-35 ppm downfield from signals for the corresponding free ligands in the 31P NMR spectra. Likewise, reaction of 1a with Ni(COD)2 yields the analogous nickel pincer complex 4a. While reaction of 1a with Pd2(dba)3 is rapid and complete in 1 h at room temperature to yield 3a, reaction of the more hindered 1c with Pd2(dba)3 requires 12 h or more for completion. These new pincer complexes are robust, display great air stability, and can be purified by flash column chromatography on silica gel under ambient conditions. Full characterization of 3a-c and 4a was possible by a combination of both 1-D and 2-D (HCOSY, HMQC and HMBC) NMR spectroscopy.10
(7) (a) Smith, R. C.; Protasiewicz, J. D. Organometallics 2004, 23, 4215. (b) Smith, R. C.; Bodner, C. R.; Earl, M. J.; Sears, N. C.; Hill, N. E.; Bishop, L. M.; Sizemore, N.; Hehemann, D. T.; Bohn, J. J.; Protasiewicz, J. D. J. Organomet. Chem. 2005, 690, 477.
(8) Vinod, T.; Hart, H. J. Org. Chem. 1990, 55, 881. (9) Selected data for 1a-c are as follows. 1a: 31P NMR (CDCl3, 162 MHz) δ -10.3, -10.2 (∼1:1). 1b: 31P NMR (CDCl3, 162 MHz) δ 1.3. 1c: 31P NMR (CDCl , 162 MHz) δ 30.4, 32.4 (∼10:1). 3
Communications
Organometallics, Vol. 25, No. 14, 2006 3303
Figure 2. Temperature-dependent 1H NMR spectra (toluene-d8, 600 MHz) and a calculated (SPARTAN PM3) space-filling molecular model of 3c (as viewed down the Br-Pd bond).
Complex 3a was further characterized by single-crystal X-ray diffraction analysis (Figure 1).11a The crystal structure of 3a reveals a four-coordinate planar palladium(II) center, as well as a crystallographically imposed C2 axis passing through the bromine, palladium, and C(1) atoms. Although individual (10) Selected data for 3a-c and 4a are as follows. 3a: 1H NMR (CDCl3, 600 MHz) δ 2.91 (doublet of virtual triplets, JHH ) 13 Hz, 2H), 3.21 (doublet of virtual triplets, JHH ) 13 Hz; 2H), 6.38 (m, 2H), 6.65 (d, JHH ) 7 Hz, 2H), 6.73 (d, JHH ) 7 Hz, 2H), 7.02 (t, JHH ) 7 Hz, 1H), 7.10-7.15 (m, 8H), 7.20 (t, JHH ) 7 Hz, 4H), 7.28 (t, JHH ) 7 Hz, 2H), 7.45 (t, JHH ) 7 Hz, 4H), 7.57 (t, JHH ) 7 Hz, 2H), 7.58-7.61 (m, 4H); 31P NMR (CDCl3, 162 MHz) δ 26.3 (s). 3b: 1H NMR (CDCl3, 400 MHz) δ 1.08-1.82 (m, 38 H), 1.97 (d, JHH ) 12 Hz, 2H), 2.13 (s, 2H), 2.79 (m, 2H), 2.53 (doublet of virtual triplets, JHH ) 13 Hz, 2H), 2.84 (doublet of virtual triplets, JHH ) 13 Hz, 2H), 6.75 (d, JHH ) 8 Hz, 2H), 7.04 (t, JHH ) 8 Hz, 1H), 7.227.25 (m, 4H), 7.27-7.31 (m, 2H), 7.34 (d, JHH ) 7 Hz, 2H); 31P NMR (CDCl3, 162 MHz) δ 33.7 (s). 3c: 1H NMR (CDCl3, 400 MHz) δ 1.29 (virtual triplet, 18H), 1.32 (broad, 6H), 1.59 (broad, 6H), 1.84 (broad, 6H), 2.69 (doublet of virtual triplets, JHH ) 13 Hz, 2H), 2.93 (doublet of virtual triplets, JHH ) 13 Hz, 2H), 6.58 (d, JHH ) 8 Hz), 6.94 (t, JHH ) 8 Hz), 7.22 (m, 2H), 7.32 (m, 4H), 7.37 (m, 2H); 31P NMR (CDCl3, 162 MHz) δ 55.1 (s). 4a: 1H NMR (CDCl3, 400 MHz) δ 2.71 (doublet of virtual triplets, JHH ) 12 Hz, 2H), 3.04 (doublet of virtual triplets, JHH ) 12 Hz, 2H), 6.26 (m, 2H), 6.58 (d, JHH ) 7 Hz, 2H), 6.68 (d, JHH ) 7 Hz, 2H), 6.94 (t, JHH ) 7 Hz, 1H), 7.07-7.18 (m, 12H), 7.25 (t, JHH ) 7 Hz, 2H), 7.46 (t, JHH ) 7 Hz, 4H), 7.57-7.65 (m, 6H); 31P NMR (CDCl3, 162 MHz) δ 27.0 (s). (11) (a) Crystallographic data for 3a: yellow blocks, crystal dimensions 0.45 × 0.30 × 0.29 mm3, space group C2/c, a ) 16.1720(7) Å, b ) 14.2258(7) Å, c ) 16.1442(8) Å, β ) 104.3920(10)°, V ) 3597.6(3) Å3, Z ) 4, Fcalcd ) 1.499 g cm-3, µ(Mo KR) ) 1.746 mm-1, data measured on a Bruker SMART 1000 CCD-based diffractometer (Mo KR radiation, λ ) 0.710 73 Å) at 300(1) K, structure solved by direct methods, 9361 reflections collected, 3160 unique reflections (Rint ) 0.0266), data/restraints/parameters 316/0/219, final R indices (I > 2σ(I)) R1 ) 0.030 and wR2 ) 0.083, R indices (all data) R1 ) 0.038 and wR2 ) 0.086, largest difference peak and hole +0.384 and -1.277 e Å-3. (b) Crystallographic data for 7a: pale brown block, crystal dimensions 0.12 × 0.20 × 0.26 mm3, space group Pbcn, a ) 14.9444(2) Å, b ) 11.6289(2) Å, c ) 19.7745(3) Å, V ) 3436.55(9) Å3, Z ) 4, Fcalcd ) 1.577 g cm-3, µ(Mo KR) ) 1.833 mm-1, data measured on a Bruker Kappa APEX II CCD-based diffractometer (Mo KR radiation, λ ) 0.710 73 Å) at 100(2) K, structure solved by direct methods, 11 657 reflections collected, 6006 unique reflections (Rint ) 0.0306), data/restraints/parameters 6006/0/235, final R indices (I > 2σ(I)) R1 ) 0.036 and wR2 ) 0.077, R indices (all data) R1 ) 0.068 and wR2 ) 0.089, largest difference peak and hole +0.679 and -0.727 e Å-3. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
molecules of 3a are chiral, the crystal is comprised of both enantiomers, which are related by a crystallographic inversion center. The bond lengths of Pd-P and Pd-C are in reasonable agreement with published values for PCP pincer complexes.12,13 In comparison to the known closely related PCP pincer complex [(C6H3(CH2PPh2)2)PdBr)] (5a),13 having P-Pd-C bond angles of 80.6 and 81.8°, complex 3a has C(1)-Pd-P bond angles of 84.74°. These values are actually closer to those of the nonpincer analogue trans-[(PPh3)2Pd(C6H5)Br]14 (86.3 and 88.8°), which does not possess any ring constraints. The most striking difference between the solid-state structures of 3a and 5a lies in the distortion of P atoms away from the plane containing the anchor atom C(1), its ring, and the Pd and Br atoms. One can assign the “twist” angle Φ (see Scheme 1), defined as the angle between the plane of the anchoring ring and square plane containing the metal and its four directly attached atoms, to assess this effect for comparative purposes. A Φ value of 76.0° for 3a greatly surpasses the Φ ) 18.4° found for 5a.13 The twist angle for 3a is also 34° greater than that found for CEC pincer complexes (C ) N-heterocyclic carbene) having the largest Φ values reported to date (up to 41.8°).4 The structural rigidity of this pincer platform is maintained up to 130 °C, as shown by variable-temperature 1H NMR studies of 3a in CDCl2CDCl2 (see the Supporting Information for details). Analysis of spectra for the tert-butyl derivative 3c (Figure 2) revealed hindered rotation about the C(tert-butyl)-P bonds, as well as an equally rigid pincer backbone. At low temperature, the two types of tert-butyl groups are inequivalent, and remain so, up to 105 °C in toluene-d8. One of the signals, however, is resolved into three independent methyl resonances at low temperature. This observation suggests that the tert-butyl (12) (a) Gorla, F.; Venanzi, L. M.; Albinati, A. Organometallics 1994, 13, 43. (b) Kraatz, H. B.; Milstein, D. J. Organomet. Chem. 1995, 488, 223. (c) Cotter, W. D.; Barbour, L.; McNamara, K. L.; Hechter, R.; Lachicotte, R. J. J. Am. Chem. Soc. 1998, 120, 11016. (d) Longmire, J. M.; Zhang, X. M.; Shang, M. Y. Organometallics 1998, 17, 4374. (13) Bachechi, F. Struct. Chem. 2003, 14, 263. (14) Flemming, J. P.; Pilon, M. C.; Borbulevitch, O. Y.; Antipin, M. Y.; Grushin, V. V. Inorg. Chim. Acta 1998, 280, 87.
3304 Organometallics, Vol. 25, No. 14, 2006 Scheme 4. Synthesis of Diphosphinite Complex 7a
groups most proximal to the bromine atom face steric clashes with the bromine atom and cause rotation about the C-CMe3 bond, thus resulting in three inequivalent methyl groups (portrayed as A, A′, and A′′ in Figure 2, right). If the atropisomerization process depicted in Scheme 1 were occurring, the two types of tert-butyl groups would undergo exchange and lead to NMR broadening. In contrast, the pincer PCP complexes [(C6H3(CH2PtBu2)2)PdCl)]1 and [(C6H3(CH2PtBu2)2)Pd(THFd8)]BF415 only show a single 1H NMR resonance for their freely rotating and equivalent tert-butyl groups. This new type of pincer architecture is expected to be very general and adaptable for a broad variety of donor atoms and for many transition metals. For example, starting with 2,6-(2HOC6H4)2C6H3Br, the new diphosphinite ligand 6a can be prepared. This material reacts with Pd2(dba)3 to yield the analogous pincer complex 7a (Scheme 4). The overall structure is quite similar to that determined for 3a (Φ ) 73.8°, dPd-P ) 2.2824(5) Å, dPd-C1 ) 2.063(3) Å).11b The other intermediates described in Scheme 3 lend themselves naturally for easy access to other types of high-twist-angle pincer ligands and complexes.16 These pincer ligands can be contrasted to the related binding mode of the 2,6-(2-MeOC6H4)2C6H3 terphenyl (Danip) (15) Kimmich, B. F. M.; Marshall, W. J.; Fagan, P. J.; Hauptman, E.; Bullock, R. M. Inorg. Chim. Acta 2002, 330, 52.
Communications
developed for use in stabilizing lanthanide complexes. While twist angles of up to 41° have been characterized, the rigidity of these systems is unknown.17 In summary, we have prepared new pincer ligands constructed using m-terphenyl scaffolds and showed their efficacy in yielding novel palladium and nickel pincer complexes. A structural analysis of 3a shows that despite a chelate ring size larger than that of earlier conventional pincer complexes (compare 7 to 5), the new pincer complexes better match the metrical parameters found for related unconstrained nonchelate structures. In addition, these systems have the greatest twist angles determined to date for pincer complexes and also display a very high degree of nonfluxionality (up to 130 °C). These air- and heat-stable materials thus hold much potential for resolution and use in catalysis. Efforts are now ongoing to resolving the enantiomers of 3a-c and to examine their catalytic behavior relative to that of conventional pincer complexes. Acknowledgment. This work was supported by the National Science Foundation (Grant No. CHE 0202040). We thank Dr. Charles Campana (Bruker AXS, Inc.) for assistance with the crystallographic determination of compound 7a. Supporting Information Available: Text and figures giving experimental procedures, NMR spectra, and analysis results and CIF files giving X-ray crystallographic data. This material is available free of charge via the Internet at http://pubs.acs.org. OM060256Z (16) A related diimine pincer complex (2,6-(2-(PhNdC(H))2C6H4)2C6H3PdI) has been structurally characterized with Φ ) 63.6°: Ma, L.; Protasiewicz, J. D. Manuscript in preparation. (17) (a) Rabe, G. W.; Zhang-Presse, M.; Riederer, F. A.; Yap, G. P. A. Inorg. Chem. 2003, 42, 3527. (b) Rabe, G. W.; Berube, C. D.; Yap, G. P. A. Inorg. Chem. 2001, 40, 4780.
